Can Data Save the Planet? Inside UVA Darden’s High-Stakes Model for Preventing Asteroid Impacts.

Some 66 million years ago, a giant asteroid 10 kilometers wide streaked into Earth’s atmosphere, crashing near Mexico’s Yucatan Peninsula and wreaking havoc of epic proportions. The impact incinerated everything within several hundred miles, triggered megatsunamis and global wildfires, and filled the air with so much debris that the sun was blocked out for years.

“It caused the extinction of 75% of the earth’s species at the time, including all of the non-avian dinosaurs,” says Asa Palley, an associate professor in the Data Analytics & Decision Sciences area at the University of Virginia Darden School of Business. “It was a global catastrophe.”

The chances of an asteroid impact like that in our lifetimes are infinitesimally small — but they are not zero. That leaves policymakers in a bind. How much should they invest in technology to detect and stop future asteroids hurtling towards Earth? And how can they even begin to make that decision?

That’s where Palley comes in. As an expert in decision science, he and his colleagues have written a paper titled “Evaluating Investments in Asteroid Detection Technologies to Prevent Catastrophic Impacts on Earth,” which has been accepted for publication in the journal Management Science.

Among other findings, they argue that it’s not planet killers like the monster that hurtled into the Earth at the end of the Cretaceous Period we should be worried about, but rather the far more numerous smaller asteroids that could cause lesser but still locally catastrophic damage, potentially killing millions of people. And the amount we should invest in stopping them may depend on how much we value those lives.

Two of Palley’s co-authors, Ralph Keeney of Duke University’s Fuqua School of Business and Victor Jose of Georgetown University’s McDonough School of Business, first raised the question more than a decade ago. The more he thought about it, the more Palley saw it as a perfect question to approach through the discipline of decision analysis, the application of mathematical models to help us make better choices.

“We make thousands of decisions every day, but most of them are not worth spending hundreds of hours analyzing using a formal mathematical model,” he says. The asteroid question, however, checked every box: a high-stakes decision without an immediate answer, involving many possible outcomes, uncertainty and risk, and multiple competing objectives.

In addition to saving lives, stopping asteroid impacts can also help prevent property damage and environmental collapse. Multiple factors also affect the severity of impact, including the size of the asteroid and the location where it hits.

In 1908, a 50- to 60-meter wide asteroid exploded over Siberia, flattening 800 square miles of forest, but thankfully costing few lives. The same asteroid hitting New York or Shanghai would have killed millions. Weighing all those factors, decision analysis doesn’t necessarily provide a simple answer.

“What it does, is provide a formal, quantitative framework to map your subjective preferences,” Palley says. “It helps you consider the trade-offs.”

To work on the problem, Palley, Jose, and Keeney recruited Palley’s brother, Thomas Palley, a postdoctoral fellow at Harvard Business School, and Mario Juric, an astronomer at the University of Washington.

Along the way, Palley had to practically become an astronomer himself, spending countless hours researching questions such as how many asteroids exist within a reasonable distance from Earth, what the probability is that we might find ourselves on a collision course with one, and what we could do if we were.

“It was fun to pretend to do astrophysics for a little bit,” Palley admits.

He and his colleagues eventually constructed a decision tree with more than 200 branches to represent all the contingencies, calculating the probability of each separately. In doing so, they discovered that the combined potential threat posed by giant asteroids of 500 meters or more is paradoxically lower than that of the smallest asteroids, 140 meters or less.

That’s in part because there are fewer large asteroids, but also because we know where almost all of them are, so it’s unlikely one would catch us unawares. By contrast, there are an estimated 150,000 smaller asteroids in our corner of the galaxy, and we know where almost none of them are. While experimental technology exists to deflect or destroy an asteroid, it would take a long time to deploy.

“If you found out an asteroid is potentially on track to hit Earth in less than a year, there’s not much you could do to prevent the impact,” Palley says. Even evacuating an area to minimize loss of life would require substantial advance notice.

To help catalog the population of those smaller asteroids and provide notice before they hit Earth, the researchers considered two advanced telescopes then under consideration, the powerful land-based Vera Rubin Observatory Large Synoptic Survey Telescope in Chile, which would cost an estimated $700 million to build and operate; and the Near-Earth Object Surveyor, a space-based telescope which would orbit Venus and would cost another $800 million.

Using a complex mathematical model, they created a cost-benefit analysis to gauge which of these investments would be worth it. Despite the many variables involved, they quickly determined that the most important factor was the valuation assigned to each human life.

While putting a price figure on human life may seem macabre, it’s a necessary component of economic analysis.

“Economics is the study of the allocation of scarce resources, and we have to make judicious decisions about which opportunities to spend our limited funds offer the greatest collective benefit,” says Palley. “In the end, what governments invest in should be aligned with some consistent assessment of what it’s worth to us as a society to make the world safer.”

In the academic literature, economists have determined the value of a statistical life by looking backwards at projects governments have been willing to invest in, seeing how many lives could be saved by them, and then considering the cost. That calculus has resulted in figures ranging from $45,000 to $18.3 million per life, depending on the country and the situation — with the United States’ National Highway Traffic Safety Administration most recently placing the “value of a statistical life” at around $14.2 million.

Based on their model, Palley and his colleagues determined that the Vera Rubin telescope, which could detect up to 90% of all asteroids, was worth the investment when the cost of a human life was set at more than $1.18 million. Adding the NEO Surveyor to the equation made sense when each human life was valued over $3 million. Thus, according to most analyses of human value, both investments ultimately seem worth it.

“It’s not our role to dictate what that number should be,” Palley stresses. “But if you think human life is more precious than a certain amount, our model is going to tell you that you should invest in this expensive satellite to accelerate the rate of discovery.”

In the years that Palley and his colleagues have worked on this project, the Vera Rubin telescope has been built and opened in 2025, and the NEO Surveyor has been proposed for deployment in 2027.

Asteroid detection is only one of Vera Rubin’s missions; however, ensuring it continues to scan for those threats would require an ongoing investment of hundreds of millions of dollars. And there are no guarantees the NEO Surveyor will be launched, or come in at its estimated price.

Palley hopes their research will help guide policymakers in these considerations and give them more confidence in their choices. If they want to make their own analysis, he and his colleagues have posted their decision tree and model online so analysts can tweak the numbers and try their own hand at the process.

“Public policy is an area where it makes sense to bring this sort of precise, quantitative analysis to fully understand big, important problems to make better decisions,” says Palley. And if the model happens to help save the world in the process, well, that’s just a bonus.

Professor Asa Palley is co-author of “Evaluating Investments in Asteroid Detection Technologies to Prevent Catastrophic Impacts on Earth,” with Thomas Palley, Victor Jose, Ralph Keeney and Mario Juric. It is forthcoming the journal Management Science.